Skip to main content
Security

Enterprise security for everyone

Container isolation with the dangerous capabilities dropped, a web application firewall, behavioural bot protection at the origin, and real-time malware scanning. No security tier, no add-ons - it is how every account is built.

14 days free. No credit card. No commitment.

Isolated

A container per account

No add-ons

Security and backups built in

99.99%

Uptime, last 12 mo

Daily

Backups kept up to 30 days

E

210+

Tasks Ellie handles

Isolation

Inside your container

Shared hosting usually means a folder with different permissions. This is the configuration each account's container is started with - the specifics, because on a security page the specifics are the argument.

  • Privileges cannot grow

    Containers start with no-new-privileges set, so a process inside can never acquire more than it began with - setuid binaries included. Whatever an exploit lands as, it stays.

  • The dangerous capability is dropped

    SYS_ADMIN - the capability that unlocks mounting, namespace manipulation and most of the interesting container-escape research - is dropped outright. What remains is the minimum the container needs to run.

  • Nothing executes from writable space

    Your home directory is mounted noexec and /tmp is a noexec tmpfs. An attacker who manages to upload a binary cannot then run it, which removes the second half of most upload-based attacks.

  • Resources are bounded per account

    Each container sits in its own systemd cgroup slice with its own CPU, memory and I/O limits. A neighbour under load, or being attacked, cannot take your share.

  • It comes back on its own

    Containers are set to restart always. A crash - yours or something we caused - is measured in seconds, without anybody needing to notice first.

Defence in depth

Three places an attack gets stopped

Layers only count if they fail independently. These do: getting past the firewall does not help you past the container, and neither helps against a stolen password.

Before it reaches your code

Behavioural bot protection

Every request is scored in real time on how it behaves rather than what it claims to be, with challenges for the suspicious and bans that propagate across the fleet. It runs at the origin, so it also catches what a CDN in front of you would have waved through.

ModSecurity with the OWASP rules

The web application firewall sits in front of your application and blocks the classics - SQL injection, cross-site scripting, traversal - before your code ever parses the request. Per-site exclusions when a rule fights your app.

Rate limiting per subdomain

Set limits per site rather than per server, with your own whitelist for the addresses that should never be throttled. This is what stops credential stuffing against a login form.

IP, range and country blocking

Block a single address, a CIDR range, or an entire country from the panel. Datacenter traffic and AI crawlers each have their own switch, because the answer for those is rarely the same as for people.

On the server

A container per account

Not a folder with different permissions. Each account is a real podman container with its own process space, its own filesystem view and its own resource ceiling - the flags are listed above.

SELinux enabled and enforcing

The hosts run with SELinux on rather than switched off to make a control panel work, which is more than most shared hosting can say. Policy exceptions exist where the stack needs them, and the isolation guarantees above do not depend on them.

Real-time file scanning

A filesystem watcher queues files for scanning as they are written, not on a nightly sweep, and known-bad patterns are flagged within seconds of appearing rather than the next morning.

WordPress vulnerability matching

Your installed plugins and themes are matched against published CVEs, so you find out that the thing you installed two years ago now has a known hole - and can update it from the same screen.

Dependency scanning for deployed apps

Every build records the packages it actually installed, and they are re-checked nightly against OSV, the public advisory database. An advisory published months after your last deploy still reaches you, with the version that clears it and whether the package serves requests or only built them.

Around your account

Passkeys, not just passwords

Sign in with a passkey - Touch ID, Windows Hello, a hardware key - and there is no shared secret left to phish or reuse. Available on your account today.

SSH keys pinned to addresses

Every key can be restricted to the IPs allowed to use it, so a stolen key is not automatically a working key. Keys are managed and revoked from the panel.

An audit log of every change

Every create, update and delete is recorded with who did it, when, and what the value was before and afterwards. Searchable, and it covers Ellie as well as people.

Delegate access without sharing a login

Give a developer, an agency or your accountant their own login instead of your password. Choose the hosting accounts they reach and set each feature to view or edit - or share only your invoices. Their actions are theirs in the audit log, and revoking them takes effect on their next click without changing your own credentials.

The layer we built ourselves

Bot protection built into the server

Most hosts buy this from a CDN. Ours scores every request at the origin, which means it also sees what a CDN passes through - and it is on for every account from the first day.

Blocklists tell you where an address has been. Behavioural scoring watches what it does on your site - the pace, the order of the paths, whether it keeps a cookie, whether it can run the challenge it was handed - and decides from that.

It is the part of the platform we have put the most work into, it has its own page, and Ellie will drive the whole thing for you in plain English.

Included, on, and not a tier

  • Every plan
  • On by default
  • Nothing to install
  • No plugin overhead
  • Per-site tuning

Straight answers

What this does not do

A security page that only lists wins is a sales page. These are the limits worth knowing before you rely on any of it.

  • Shared hosting is shared

    Your files and processes are isolated. The machine underneath is not exclusive to you. If you need single-tenant hardware, that is a different product from a different kind of provider.

  • Your application is still yours

    We can block the request that carries an attack and refuse the binary that arrives with it. We cannot fix an unpatched plugin or a leaked admin password. Most compromises we see arrive through the application, logged in and looking legitimate.

  • Backups are the last layer, and they are on

    Everything above reduces the chance of a bad day. Daily backups on independent servers are what makes a bad day survivable, which is why they are on every plan rather than sold as an add-on.

Pricing

Simple, transparent pricing

Every plan includes managed WordPress, SSH access, daily backups, and enterprise-grade security. Start with a 14-day free trial.

Black Friday80% off your first year with code BLACKFRIDAY80Ends Cyber Monday, November 30

80% OFF

Startup

Great for growing businesses

$7.92

$1.58/mo

$19.00 for your first year, then $94.99/yr

  • 1 website
  • 10 GB storage
  • ~10,000 visits/month
  • 5 MySQL databases
  • 250,000 inodes
  • 5 FTP users

What's included:

SSL & Domain

  • Free SSL certificate
  • Free starter domain

Email

  • 2 mailboxes
  • 3 GB per mailbox
  • Webmail, IMAP and SMTP
Most popular
80% OFF

Advanced

For professional websites

$15.00

$3.00/mo

$36.00 for your first year, then $179.99/yr

  • 5 websites
  • 20 GB storage
  • ~110,000 visits/month
  • 10 MySQL databases
  • 500,000 inodes
  • 10 FTP users

Everything in Startup, plus:

Email

  • 5 mailboxes
  • 5 GB per mailbox

Backups

  • 14 days of backup history

Performance

  • Memcached object cache
  • Redis object cache
80% OFF

Pro

Maximum performance and features

$22.92

$4.58/mo

$55.00 for your first year, then $274.99/yr

  • 10 websites
  • 40 GB storage
  • ~200,000 visits/month
  • 20 MySQL databases
  • 750,000 inodes
  • 20 FTP users

Everything in Advanced, plus:

Email

  • 10 mailboxes
  • 10 GB per mailbox

Deployment

  • 2 SSR applications

Backups

  • 30 days of backup history

Monitoring

  • 1 minute between checks

For new customers on annual plans. Enter the code at checkout, or when you upgrade a free trial before the sale ends. The 80% comes off your first year; renewals are at the plan's regular price. Ends 11:59 p.m. Eastern on Monday, November 30.

Security included

Every plan, every layer

Container isolation, the firewall, bot protection, malware scanning, SSL and daily backups are in every plan. There is no security tier to upgrade to, because there is nothing held back.

Questions

Frequently asked questions

What separates my account from the one next to it?
A container, not a folder. Each account runs as its own podman container with a separate process space, its own filesystem view, and its own CPU, memory and I/O ceiling enforced by a systemd cgroup slice. Inside it, privileges cannot escalate, SYS_ADMIN is dropped, your home directory and /tmp cannot execute anything.
Do you run SELinux?
Yes, enabled and enforcing on every host - which is worth saying because a lot of shared hosting cannot, since the popular control panels require it to be switched off. It is one layer among several rather than the thing everything else depends on: the isolation guarantees come from the container configuration, which is listed in full above.
How does the real-time malware detection differ from a nightly scan?
A filesystem watcher sees the write as it happens and queues the file immediately, so something dropped at 3pm is flagged at 3pm rather than at 3am the next morning. That gap is the whole point - a nightly scanner gives an attacker most of a day to use what they uploaded.
How is the dependency scanning different from running npm audit myself?
Three ways that matter. It reads the tree that was actually installed rather than your manifest, so it reports the versions running and not the ones a version range allows. It re-checks nightly against OSV, so an advisory published after your last build reaches you without a rebuild, which is the case an install-time check can never catch. And it covers npm and RubyGems through one matcher. You still get the npm output in your build log; this is the part that keeps working after the build finishes.
What do you do about a vulnerable package in my app?
Tell you, clearly, and nothing else. Upgrading a package is a commit in your repository and we build what you push rather than editing your source. Findings are grouped by package with the single version that clears them, production dependencies are separated from build-only ones, and a weekly email goes out only when something changes: a new critical or high finding, one that just became fixable, or an open one nobody has looked at in a month. Ellie can answer questions about it too.
Is the bot protection the same thing as the firewall?
No, and they catch different things. The WAF matches request content against known attack patterns. The bot protection scores behaviour over time - how fast, in what order, from where, with what fingerprint - which is what catches the traffic that looks perfectly legitimate one request at a time. Both run on every plan.
What happens if my site is compromised anyway?
Restore from a backup, which every plan has daily and which lives on servers independent of your hosting. You can restore a single file, a single database or the whole account. Then work out how they got in - the audit log and the access logs are both there, and we will help you read them.
Can I lock the panel down further?
Yes. Sign in with a passkey instead of a password, restrict each SSH key to specific IP addresses, and put a site behind HTTP password protection. For collaborators, delegate access gives them their own login instead of your credentials, limited to the hosting accounts and features you choose, view or edit. It is revocable in one click, and their changes are attributed to them in the audit log.
Do I pay extra for any of this?
No. There is no security tier and no add-on. Container isolation, the WAF, bot protection, malware scanning, SSL, and daily backups are what the platform is, not what it upsells.

Recent articles